IMPLEMENTASI ALGORITMA LOGISTIC REGRESSION UNTUK DETEKSI LINK PHISHING

Hilya Anbiyani Fitri Muhyidin, - (2024) IMPLEMENTASI ALGORITMA LOGISTIC REGRESSION UNTUK DETEKSI LINK PHISHING. S1 thesis, Universitas Pendidikan Indonesia.

Abstract

Peningkatan pengguna internet yang massif memicu peningkatan tindak kejahatan dunia maya melalui jaringan internet salah satunya phishing yang mana berpotensi
merugikan pihak yang terkena dampak dalam hal finansial. Tujuan dari penelitian ini melengkapi penelitian terdahulu yaitu membuat website aplikasi deteksi link phsihing dengan menerapkan algoritma logistic regression sehingga diharapkan dapat mengurangi resiko dari kejahatan phishing. Metode dan desain penelitian menerapkan Framework AI
Project Cycle untuk proses pengembangan website aplikasi. Dataset yang digunakan merupakan data URL yang diperoleh dari berbagai sumber. Berdasarkan hasil penelitian dapat disimpulkan: 1) Model logistic regression diimplementasikan ke dalam bentuk website aplikasi deteksi link phishing melalui tahapan Framework AI Project Cycle dengan memanfaatkan teknologi HTML, CSS, JavaScript, Flask framework, dan ngrok; 2) Algoritma model logistic regression yang dilatih menggunakan fitur top-6 dan fitur tambahan path_len menghasilkan skor akurasi sebesar 90,21% setelah dilakukan 10-fold cross validation, serta memiliki skor rata-rata precision sebesar 90%, recall sebesar 91%, dan f1-score sebesar 90%.
-----
The massive increase in internet users has triggered a rise in cybercrime through the internet, one of which is phishing, which has the potential to cause financial losses to affected parties. The aim of this research is to build upon previous studies by developing a website application for phishing link detection using the logistic regression algorithm, with the hope of reducing the risks of phishing crimes. The research method and design utilize the AI Project Cycle Framework for the website application development process. The dataset used comprises URL data obtained from various sources. Based on the research results, the following conclusions can be drawn: 1) The logistic regression model was implemented in a website application for phishing link detection through the stages of the AI Project Cycle Framework, utilizing HTML, CSS, JavaScript, Flask framework, and ngrok technologies; 2) The logistic regression model algorithm, trained using the top-6 features and an additional path_len feature, achieved an accuracy score of 90.21% after 10-fold cross-validation, with an average precision score of 90%, recall score of 91%, and
f1-score of 90%.

	Text S_PSTI_2005517_Title.pdf Download (855kB)
	Text S_PSTI_2005517_Chapter1.pdf Download (300kB)
	Text S_PSTI_2005517_Chapter2.pdf Restricted to Staf Perpustakaan Download (634kB)
	Text S_PSTI_2005517_Chapter3.pdf Download (631kB)
	Text S_PSTI_2005517_Chapter4.pdf Restricted to Staf Perpustakaan Download (1MB)
	Text S_PSTI_2005517_Chapter5.pdf Download (233kB)
	Text S_PSTI_2005517_Appendix.pdf Restricted to Staf Perpustakaan Download (1MB)

Item Type:	Thesis (S1)
Additional Information:	https://scholar.google.com/citations?hl=en&user=a7ryywQAAAAJ ID SINTA Dosen Pembimbing: Suprih Widodo: 5978120 Liptia Venica: 6779029
Uncontrolled Keywords:	AI Project Cycle, Logistic Regression, Machine Learning, Phishing Link Detection. AI Project Cycle, Logistic Regression, Machine Learning, Phishing Link Detection.
Subjects:	L Education > L Education (General) T Technology > T Technology (General)
Divisions:	UPI Kampus Purwakarta > S1 Pendidikan Sistem Teknologi dan Informasi
Depositing User:	Hilya Anbiyani Fitri Muhyidin
Date Deposited:	11 Jul 2024 03:41
Last Modified:	11 Jul 2024 03:41
URI:	http://repository.upi.edu/id/eprint/118713

Actions (login required)

View Item